© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 39309
    Prev
    Next
    INDEX
    Explanations

    medical and proper nouns

    np_acts-logits-general · gemini-2.5-flash-lite

    document sections containing metadata like dates, timestamps, or formatting breaks in web content.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron fires most strongly on blank‐line or paragraph‐break tokens, i.e. spots in the text where there are extra newlines or section separators.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     bleus
    -1.75
     dentelle
    -1.73
     brune
    -1.65
     récents
    -1.65
     eene
    -1.65
     esqueleto
    -1.62
     blz
    -1.62
     indisponible
    -1.60
     paillettes
    -1.60
     mengumumkan
    -1.59
    POSITIVE LOGITS
    ’,
    1.91
    ↵↵
    1.88
    ’
    1.76
     we
    1.69
     ваше
    1.66
    gouv
    1.65
    但在
    1.61
    ;;)
    1.59
    on
    1.58
     replete
    1.57
    Activations Density 0.000%

    No Known Activations